J. Med. Chem. 1996, 39, 1361-1371
1361
Articles Somatostatin Receptor-Binding Peptides Labeled with Technetium-99m: Chemistry and Initial Biological Studies Daniel A. Pearson,* John Lister-James, William J. McBride, David M. Wilson, Lawrence J. Martel, Edgar R. Civitello, John E. Taylor,† Brian R. Moyer, and Richard T. Dean Department of Chemistry, Diatide, Inc., 9 Delta Drive, Londonderry, New Hampshire 03053, and Biomeasure Inc., 27 Maple Street, Milford, Massachusetts 01757 Received February 13, 1995X
The synthesis of peptides which possess a high affinity for the somatostatin receptor and contain a chelator for the radionuclide technetium-99m is described. The target compounds were designed such that they would form stable, oxotechnetium(V) chelate complexes in which the site of metal coordination was well defined and remote from the receptor-binding region. Oxorhenium(V) chelate complexes of these peptides were prepared as nonradioactive surrogates for the technetium complexes. Peptide oxorhenium complexes and Tc-99m complexes eluted closely upon HPLC analysis. The receptor-binding affinities of both the free and rheniumcoordinated species were measured in vitro. The binding affinities of the free peptides (Ki’s in the 0.25-10 nM range) compared favorably with [DTPA]octreotide (Ki ) 1.6 nM), which, as the indium-111 complex, is already approved for somatostatin receptor (SSTR)-expressing tumor imaging in the United States and Europe. Furthermore, the rhenium-coordinated peptides had binding affinities which, in many cases, were higher than those of the corresponding free peptides, with several complexes having a Ki’s of 0.1 nM. Some of the more potent SSTRbinding peptides were labeled with technetium-99m and assessed in an in vivo study with tumor-bearing rats. The 99mTc-labeled peptides prepared in this study should be useful as SSTR-expressing tumor-imaging agents due to their high SSTR-binding affinities, ease of preparation, and, because they are low molecular weight peptides, expected pharmacokinetics characterized by rapid tracer excretion from the body resulting in high-contrast images. Introduction Somatostatin (somatotropin release-inhibiting factor, SRIF) is a cyclic peptide which was initially isolated from the hypothalamus1 and has been shown to have an inhibitory effect on the secretion of many hormones, including growth hormone. Since the initial discovery, several related somatostatin peptides have been identified, with the tetradecapeptide compound 1 designated as somatostatin 14 (SRIF-14). These peptides are widely distributed throughout the body and are found in the gut, various exocrine and endocrine glands, and most organs.2 There are at least five subtypes of the SRIF receptor (somatostatin-type receptor or SSTR), and subtypes SSTR1, SSTR2, SSTR3, SSTR4, and SSTR5 have been cloned.3 Most tumors of neuroendocrine origin express receptors for SRIF to a much greater extent than normal tissue.4,5 To the extent that it has been examined, the SSTRs on tumor cells belong predominantly to the SSTR2 subtype.3c The list of the types of tumors found to express SSTRs currently includes tumors of the amine-precursor-uptake-anddecarboxylation (APUD) cell system including small cell lung carcinoma (accounting for 25% of all malignant lung cancers) (57%), endocrine pancreatic tumors (89%), metastatic carcinoids (87%), GH-producing pituitary adenomas (98%), paragangliomas (92%), and also cer* To whom correspondence should be addressed at Diatide, Inc., 9 Delta Dr., Londonderry, NH 03053. † Biomeasure Inc., 27 Maple St., Milford, MA 01757. X Abstract published in Advance ACS Abstracts, February 1, 1996.
0022-2623/96/1839-1361$12.00/0
tain breast tumors (20% of all), lymphomas (87%), astrocytomas (82%), meningiomas (98%), and some colorectal cancers (12%). Values given in parentheses are percentages of tumor samples tested and found to be SSTR positive.5 Although it has been shown to have an inhibitory effect on various tumors, the use of 1 for the treatment of cancer is hampered by its short in vivo half-life of about 3 min.6
Analogs of somatostatin have been synthesized which incorporate D-amino acids to prolong in vivo half-life by inhibiting the action of amino- and carboxypeptidases. Somatuline (BIM-23014C, compound 2)7 is a cyclic octapeptide which has been shown to inhibit the growth of tumors of the human small cell lung carcinoma (SCLC) cell line NCI-H69 in an animal model. It has been approved for use in France and is currently in phase III clinical trials in the United States. Octreotide (compound 3) is another cyclic octapeptide which has been shown to be 2000 times more effective than 1 in the suppression of growth hormone secretion in the rat 1 h postadministration.8 Compounds 2 and 3 both bind preferentially to the SSTR2 receptor subtype.3c Derivatives of octreotide have been labeled with the γ-emitting radionuclides 123I ([123I]Tyr3-octreotide) and 111In ([111In][DTPA]octreotide 4), and these radiotracers have been successfully used to detect somatostatin receptorpositive tumors by γ scintigraphy.9a © 1996 American Chemical Society
1362
Journal of Medicinal Chemistry, 1996, Vol. 39, No. 7
Pearson et al.
Scheme 1. Preparation of Cyclic Receptor-Binding Synthon
The 111In-labeled agent 4 has been approved in the United States and Europe and is now marketed under the trade name Octreoscan. Despite the success of Octreoscan as an imaging agent,9b a somatostatin analog incorporating the radionuclide 99mTc would be more desirable. As an isotope for radioisotopic imaging, 99mTc is preferred over 111In because it yields greater photon flux per unit of radiation dose delivered to the patient. It also has a convenient 6 h physical half-life (t1/2(111In) ) 60 h), is relatively inexpensive, and is available 24 h a day as a solution of [99mTc]pertechnetate (99mTcO4) in normal saline from an in-house 99Mo/99mTc generator. Although several attempts have been made to label somatostatin analogs with 99mTc, a successful clinical candidate has not been produced to date. The ideal agent would possess a chelator for technetium covalently attached to an SSTR-binding compound. Furthermore, this agent would be stable to the chemistry involved in its chelation of 99mTc using an instant kit formulation. The use of SRIF analogs which are cyclized via a disulfide bond is problematic in this regard. 99mTc is normally available in the 7+ oxidation state as 99mTcO -. In order to form a stable chelate complex, 4 the technetium must be reduced to the 5+ oxidation state, and stannous ion is normally used as the reductant. However, stannous ion can also reduce disulfide bonds10 and consequently may severely reduce the SSTR-binding affinity of disulfide-cyclized molecules. We chose to address this problem by designing molecules incorporating high-affinity SSTR-binding peptides not possessing a reductively labile disulfide bond and possessing peptide sequences which form effective technetium chelators.11 Because all nuclides of technetium are radioactive, oxorhenium complexes of these chelators were prepared to serve as substitutes for technetium complexes when assessing the binding affinity of analogs in an in vitro assay. It should be mentioned that the oxorhenium complexes of these peptides (when in the radioactive 186Re or 188Re form) may be useful for the radiotherapy of SSTR-expressing tumors. It has been shown that the residues Phe-D-Trp-LysThr of 2 and 3 closely approximate the key binding elements in the 7-10 segment of the receptor-bound hormone.12 Cyclic analogs which incorporate this pharmacophore for somatostatin receptors have been prepared in which the ring is formed in a “head-to-tail” fashion where the N-terminal amino group is linked to the C-terminal carboxyl group via an amide bond. One such analog is the very high affinity SSTR-binding cyclic peptide MK-678 (compound 5).13,14 Recognizing that this approach to constraining the pharmacophore in a nonreducible cyclic structure fit our needs, we synthesized a novel modification of this structure which
incorporated a site for attaching a chelator for 99mTc (or Re) while maintaining the potent biological activity of these analogs. The details of their synthesis and biological activity in vitro and in vivo are described herein.
Chemistry The cyclic receptor-binding portion was synthesized in solution as outlined in Scheme 1. Starting with valine, methyl ester, the amino acids were added sequentially as their succinimide esters to the free amine of the growing peptide chain. The intermediates were carried through without purification up to the tetrapeptide 8, which was purified by silica gel flash chromatography. The synthesis was continued as before, with BOP-Cl reagent15 used for the difficult coupling of Fmoc-hCys(Trt)-OH to the N-terminal Nmethylphenylalanine. Once the N-terminal Fmoc group and the C-terminal methyl ester were removed, the resulting crude hexapeptide was cyclized using HBTU reagent.16 Finally the side chain-protecting groups were removed with trifluoroacetic acid using water, ethanedithiol, and triisopropylsilane17 as scavengers to yield 13, which was purified by preparative reversed-phase HPLC. Peptides representing the portion of the molecule designed to serve as a chelator were synthesized by a standard solid-phase protocol18 using Rink amide resin19 as the support. All couplings were performed with HBTU reagent. In those sequences which did not contain an arginine, the cysteine sulfhydryl was protected with a trityl group. These peptides were cleaved from the resin using TFA/water (95:5) with simultaneous removal of the tert-butyl-based protecting groups
99mTc-Labeled
Somatostatin Receptor-Binding Peptides
Table 1. Chloroacetyl Intermediates from Solid-Phase Peptide Synthesis
peptide
compd
-Cys(Trt)-Gly-Cys(Trt)-OH -Cys(Trt)-Gly-Cys(Trt)-NH2 -(-Lys)-Gly-Cys(Trt)-NH2 -Lys-(-Lys)-Gly-Cys(Trt)-NH2 -Gly-Gly-Cys(Trt)-Lys-NH2 -(-Lys)-Gly-Cys(Trt)-Lys-NH2 -Gly-Gly-Cys(Mob)-Arg-NH2 -(-Lys)-Lys-Cys(Trt)-NH2 -Gly-Gly-Cys(Trt)-Lys-Lys-NH2 -Gly-Gly-Cys(Trt)-Orn-NH2 -Gly-Gly-Cys(Trt)-Orn-Asp-Orn-NH2 -(-Lys)-Gly-Cys(Trt)-Lys-Lys-NH2 -Lys-Lys-Cys(Trt)-NH2 -Lys-Lys-Cys(Trt)-Lys-NH2 -Gly-Gly-Cys(Trt)-Lys-Lys-Lys-NH2 -Gly-Gly-Cys(Mob)-Arg-Arg-NH2 -Gly-Gly-Cys(Mob)-Arg-Lys-NH2 -Gly-Gly-Cys(Mob)-Arg-Asp-NH2 -Gly-Gly-Cys(Trt)-Orn-Asp-NH2 -Gly-Gly-Cys(Trt)-Lys-Asp-Lys-Asp-NH2 -(δ-Orn)-Gly-Cys(Trt)-Lys-NH2 -(β-Dap)-Lys-Cys(Trt)-Lys-NH2
14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
and, by adapting the workup conditions, with selective reprotection of the thiol group(s). This was accomplished by removing the TFA of the cleavage mixture in vacuo followed by dissolution of the crude peptide in chloroform, which was also removed in vacuo. After several such treatments with chloroform, all of the residual acid was removed with concomitant reattachment of the trityl group to the cysteine SH. This strategy did not work well when Pmc-protected arginine was present in the sequence. In these cases the cysteine SH was 4-methoxybenzyl (Mob) protected and the arginine-containing peptides were cleaved from the resin using TFA/water/thioanisole/ethanedithiol/triethylsilane (95:5:5:2.5:2), which removed all protecting groups except the S-Mob group. The above strategies were employed to produce compounds 14-35 (Table 1). These peptides, which were capped with a chloroacetyl group at the N-terminus, were coupled to 13 in a carbonate buffer (pH ) 10) as shown in Scheme 2. This reaction produced pharmacophore-chelator conjugates in which the cysteines of the chelator portion were either trityl or methoxybenzyl protected. For those compounds which were S-trityl protected, the trityl group was removed with TFA/water/ triisopropylsilane (10:0.5:0.3). For those compounds which were S-Mob protected, the Mob group was removed with TFA/BF3‚OEt2/m-cresol (10:1:1). The final products 36-57 were purified by reversed-phase preparative chromatography (g95% pure) and confirmed by mass spectral analysis. Oxorhenium(5+) complexes were prepared by reacting peptides 36-57 with Bu4NReOBr420 in DMF to produce compounds 5879. The resulting compounds were also purified by reversed-phase chromatography (g95% pure) and found to be different, with longer analytical HPLC retention times, from the uncomplexed peptides by coinjection. All rhenium complexes were also confirmed by mass spectral analysis. The technetium-99m complexes of compounds 40-43, 45, and 57 were also prepared by reacting these peptides with [99mTc]glucoheptonate in a buffered solution. The radiochemical purity of these
Journal of Medicinal Chemistry, 1996, Vol. 39, No. 7 1363
Scheme 2. Preparation of SSTR-Binding Analogs Containing a Chelator
complexes was measured by analytical HPLC. The peptide-rhenium complexes had retention times which were very close to those of the technetium complexes, thus supporting the use of rhenium complexes as surrogates for the Tc-99m complexes. Biology Peptides 4, 4 (111In complex), 36-44, 56, 57, and rhenium complexes 58-79 were tested for SSTR-binding affinities using AR42J cells (see the Experimental Section for details), which express mostly SSTR2 receptors. Compounds were assayed by incubating equal aliquots of cell membrane with [125I][Tyr-11]SRIF-14 and the compound to be assayed at a final concentration of 10-11-10-6 M. All compounds were tested in duplicate at five different concentrations. SSTR binding was assessed from determinations of free vs cell membranebound radioactivity. Analysis of the data gave inhibition constants (Ki) via Hill plots.21 The results are presented in Table 2. A comparison was also made of compounds 40 and 42 with BIM-23014C and MK-678 in a receptor subtype specific assay. Cell membrane preparations were prepared by homogenization of the appropriate cells (human SSTR1, mouse SSTR3, human SSTR4 transfected into CHO-DG44 cells) in buffer (see the Experimental Section for details). The assays were again performed by incubating equal aliquots of cell membrane with [125I][Tyr-11]SRIF-14 and the compound to be assayed as described above. The results are presented in Table 3. The technetium-99m complexes of compounds 40-43, 45, and 57 were also administered to tumor-bearing rats and the resulting tumor uptake and biodistribution measured and compared to the 111In complex of compound 4. The results are presented in Table 4. Results and Discussion Our goal in this work was to synthesize 99mTc-labeled compounds which had high affinity for somatostatin receptors. These analogs were designed to contain a chelator for the radionuclide 99mTc and designed such that metal coordination could be accomplished easily
1364
Journal of Medicinal Chemistry, 1996, Vol. 39, No. 7
Pearson et al.
Table 2. Binding Affinities of SST Analogs and Corresponding Rhenium Complexes
for ornithine or β-amine for diaminopropionic acid) is used to form an amide bond to the next amino acid (or other carboxy-containing moiety) in the sequence. This system is similar to the triamide thiols above in that the metal chelator contains three nitrogens and one sulfur. However, unlike the triamide thiols, one of the nitrogens is a less readily ionizable primary amine. We expect metal coordination by this amine to be through its lone pair electrons, therefore forming neutral oxotechnetium(5+) complexes as shown in Figure 1. The use of the strategy described above insured that the chelator could be incorporated unambiguously at known positions during synthesis so that the effect of the chelated metal on the binding affinity of the molecule could be altered in a rational manner. In vitro SSTR binding results are shown in Table 2. Most compounds had high affinity to the SSTR, with inhibition constants (Ki’s) in the low-nanomolar or subnanomolar range. This compares favorably with [DTPA]octreotide 4 (Ki ) 1.6 nM) and [111In][DTPA]octreotide (Ki ) 1.2 nM). Changes in the distance of the chelated metal from the cyclic bioactive core seemed to have little effect on in vitro activity. Interestingly, for our series of compounds, in almost every case examined the rhenium complex had a higher binding affinity than the uncoordinated species. An exception to this phenomenon was compound 36 where the affinity of the coordinated species was 1 order of magnitude less than that of the parent compound. This may be due to the negative charge of the carboxy-terminal cysteine. Compounds 53-55, which contain a negative charge in the form of an aspartic acid side chain carboxyl, also exhibit decreased binding in relation to the rest of the compounds in this family. Whether the chelated metal was anionic or neutral also appeared to have some effect on this phenomenon. The affinity for the complexed species was much greater than that of the free peptide for compounds 41, 43, 56, and 57, which we had postulated would form neutral complexes. This discovery provides an advantage in that the affinity of the radiolabeled imaging agent can be higher than that of the excess unlabeled peptide (the molar ratio of carrier peptide to radiolabeled peptide is usually g10000:1). This produces an effectively higher specific activity of the radiolabeled imaging agent than would normally be the case and means that the unlabeled peptide should be at a competitive disadvantage with the radiolabeled peptide for the SSTRs. Compounds 40 (P587) and 42 were also examined in receptor subtype specific assays, and the binding was compared to that of the potent SSTR binders BIM23014C (2) and MK-678 (5). As seen in Table 3, both 2 and 5 were very potent for SSTR2 receptors and moderately potent for SSTR3 receptors. The affinity of these compounds was markedly decreased for both SSTR1 and SSTR4 receptors. Compounds 40 and 42 showed a much broader range of activity although they had higher affinity for the SSTR2 receptor. Once it was determined that negative charge in the region of the chelator had a deleterious effect on receptor-binding affinity, changes were made in the location of positively charged residues in an attempt to alter the biodistribution of 99mTc-labeled peptides. Technetium-99m complexes of compounds 40-43, 45, and 57 and the 111In complex of compound 4 were admin-
Ki compd (nM)a
peptide -Cys-Gly-Cys-OH -Cys-Gly-Cys-NH2 -(-Lys)-Gly-Cys-NH2 -Lys-(-Lys)-Gly-Cys-NH2 -Gly-Gly-Cys-Lys-NH2 (P587) -(-Lys)-Gly-Cys-Lys-NH2 -Gly-Gly-Cys-Arg-NH2 (P617) -(-Lys)-Lys-Cys-NH2 -Gly-Gly-Cys-Lys-Lys-NH2 -Gly-Gly-Cys-Orn-NH2 -Gly-Gly-Cys-Orn-Asp-Orn-NH2 -(-Lys)-Gly-Cys-Lys-Lys-NH2 -Lys-Lys-Cys-NH2 -Lys-Lys-Cys-Lys-NH2 -Gly-Gly-Cys-Lys-Lys-Lys-NH2 -Gly-Gly-Cys-Arg-Arg-NH2 -Gly-Gly-Cys-Arg-Lys-NH2 -Gly-Gly-Cys-Arg-Asp-NH2 -Gly-Gly-Cys-Orn-Asp-NH2 -Gly-Gly-Cys-Lys-Asp-Lys-Asp-NH2 -(δ-Orn)-Gly-Cys-Lys-NH2 -(β-Dap)-Lys-Cys-Lys-NH2 (P829) [DTPA]octreotide
36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 4
ReO complex
Ki (nM)a
1.8 1.5 2.0 0.7 2.5 4.2 0.3 2.2 0.3
58 20 59 1.7 60 0.9 61 0.5 62 0.2 63 0.3 64 0.2 65 0.4 66 0.1 67 0.2 68 0.6 69 0.5 70 0.5 71 0.3 72 0.3 73 0.1 74 0.1 75 1.7 76 2.1 77 1.8 10 78 0.1 10 79 0.3 1.6 111In complex 1.2
a Binding affinity for the SSTRs on AR42J cells using labeled SRIF-14 as ligand.
125I-
Table 3. SSTR Subtype Comparison in vitro binding affinities: Ki (nM) receptor
BIM-23014C (2)
MK-678 (5)
SSTR1 SSTR2 SSTR3 SSTR4 SSTR5 µ-opioid
789 0.3 5.6 >1000 0.10 2.2
>1000 0.2 12 >1000 5.5 >10000
compd 40 (P587) 19 1.0 89 68 >10000
compd 42 (P617) 79 0.9 152 36 5425
using a standard kit formulation.22 Furthermore, SSTRbinding affinity of the pharmacophore would not be affected by the radiolabeling process. The bioactive core chosen for this work was a cyclic hexapeptide which did not contain any reducible disulfide bonds. A convergent synthetic strategy was employed where this cyclic portion was attached, via sulfhydryl alkylation, to several different cysteine-protected peptide chelators containing a chloroacetyl moiety. This efficient reaction produced, after cysteine deprotection, analogs 36-57. Because only radioactive nuclides of technetium exist, rhenium, which is very similar to technetium in its (5+) oxo coordination chemistry,23 was used in the preparation of metal complexes to serve as surrogates for the corresponding technetium complexes in in vitro assays. Several different chelating systems were examined. These chelators included (see Figure 1) bisamide bisthiols, such as are provided by the sequence -Cys-Gly-Cys(compounds 36, 37), and triamide thiols, such as are provided by the sequence -Gly-Gly-Cys- (compounds 40 (P587), 42, 44-46, 48-55). Both of these types of chelators are expected to form anionic oxotechnetium (5+) complexes.24 A novel diamide-amine-thiol chelator system of the type -(-Lys)-Gly-Cys- was also prepared (compounds 38, 39, 41, 43, 47, 56, 57 (P829)). In this system it is postulated that the R-amine of lysine (or ornithine or diaminopropionic acid) becomes part of the coordination complex, while the -amine (δ-amine
99mTc-Labeled
Somatostatin Receptor-Binding Peptides
Journal of Medicinal Chemistry, 1996, Vol. 39, No. 7 1365
Table 4. Biodistribution of Technetium-99m Complexes % ID/g
% ID
compda
RCPb (%)
tRc (min)
tumord
blood
kidneys
liver
GIe
tumor/bloodf
tumor/muscleg
40 (3) (P587) 41 (3)
99
15.0
99
14.7
42 (3) (P617) 43 (3)
99
15.1
99
14.5
45 (3)
99
7.0
57 (3) (P829)
99
7.0
2.75 (0.50) 1.28 (0.36) 3.54 (0.63) 0.57 (0.04) 1.48 (0.70) 4.88 (0.36) 2.43 (0.74)
0.50 (0.16) 0.28 (0.08) 0.28 (0.03) 0.37 (0.57) 0.60 (0.06) 0.29 (0.02) 0.11 (0.04)
21.6 (1.7) 16.9 (4.1) 14.1 (2.6) 12.1 (7.2) 20.4 (0.3) 32.7 (2.5) 5.3 (0.3)
8.3 (0.3) 16.0 (1.1) 9.3 (0.8) 14.5 (6.2) 9.2 (0.6) 10.0 (0.1) 1.0 (0.1)
27.9 (5.2) 15.9 (1.6) 27.3 (1.3) 16.3 (3.9) 38.1 (0.4) 8.4 (0.5) 6.0 (1.2)
5.6 (0.9) 4.6 (0.2) 11.1 (0.7) 1.6 (0.2) 2.4 (1.0) 21.0 (11.1) 22.0 (5.9)
30.2 (2.7) 13.8 (1.1) 40.3 (1.1) 4.9 (0.4) 18.6 (7.0) 67.9 (26.4) 73.0 (15.2)
4 (111In complex) (9)
a Data refer to the 99mTc complex of the indicated compound number. Number of rats per study in parentheses. Letter + numbers in parentheses refer to potential clinical candidates (Diatide compound code). b Radiochemical purity of 99mTc complex as measured by HPLC (see Experimental Section). c Analytical HPLC retention time of 99mTc complex. Compounds 40-43 were run at 0-100% B/A over 20 min using a Waters Delta-Pak C18 column, 5 µm, 39 × 150 mm. Compounds 45 and 57 were run at 0-100% B/A over 10 min using a Waters Nova-Pak Radial Compression C18 column, 4 µm, 8 × 100 mm (A ) 0.1% TFA in water, B ) 0.1% TFA in 90% acetonitrile/water). d Percent injected dose contained in tumor (see Experimental Section). Standard deviation values appear beneath in parentheses. e Percent injected dose contained in gastrointestinal tract (stomach, without contents; small intestine, duodenum, jejunum, and ileum, all with contents; large intestine, cecum and colon, all with contents). f Tumor to blood ratio based on % ID/g values at 90 min. g Tumor to muscle ratio based on % ID/g values at 90 min. Nontarget values were obtained from the contralateral leg muscle.
complex surrogates and found, in most cases, to have even higher receptor-binding affinity than their parent peptides. The 99mTc complexes of the peptides described hold promise to serve as useful SSTR-expressing tumorimaging agents due to their high receptor-binding affinity, ease in preparation, and, because they are low molecular weight peptides, expected rapid pharmacokinetics. Several 99mTc-labeled peptides were administered to tumor-bearing rats and the tumor uptake and biodistribution of these peptides measured. It was found that these 99mTc-labeled peptides localized in tumors. Compound 57 (P829) displayed a biodistribution similar to that of 111In-labeled octreotide and was selected for investigation in clinical studies. Experimental Section
Figure 1. Peptidyl-chelating systems.
istered to tumor-bearing rats, and their tumor uptake and biodistribution were measured. As illustrated in Table 4, the location of the positively charged residues lysine or arginine and the proximity of the technetium complex to the cyclic pharmacophore had a pronounced effect on biodistribution. Several of the more potent compounds had high gastrointestinal uptake compared to [111In]Octreoscan. Compound 57 (P829), whose tumor uptake and biodistribution was similar to the 111In complex of compound 4, was chosen for clinical evaluation. Conclusions We designed and synthesized high-affinity SSTRbinding peptides containing sequences which form strong chelates with 99mTc. The peptides synthesized were designed such that stable, metal-coordinated derivatives could be prepared where the site of metal coordination was well defined and the SSTR-binding affinity of the peptide was not compromised. Rhenium complexes of these compounds were prepared as 99mTc
Symbols and abbreviations generally follow the IUPAC-IUB recommendations as published in Int. J. Pept. Protein Res. (1984, 24, 9-37). Amino acids used were of the L-configuration unless stated otherwise. All protected amino acids used (except N-Fmoc-S-tritylhomocysteine and N-R-Boc-N-β-Fmocdiaminopropionic acid) were purchased from either Bachem, CA, Novabiochem, or Advanced ChemTech and used as is. N-RBoc-N-β-Fmoc-diaminopropionic acid (Dap) was prepared according to literature precedent.26 N-Fmoc-S-tritylhomocysteine was prepared as indicated below. Other chemicals and solvents were purchased from either J.T. Baker, Advanced ChemTech, or Aldrich and used as is. Aldrich SureSeal solvents were used when anhydrous conditions were necessary. When indicated, an automated Applied Biosystems 431A peptide synthesizer was used. For manual peptide synthesis, the reaction vessel used was purchased from Safe-Lab, Santee, CA. Analytical HPLCs were performed on a Waters system using a Delta-Pak C18 column (300 Å, 5 µm, 3.9 × 150 mm) at a flow rate of 1.2 mL/min or on a Waters system using a radial compression Nova-Pak C18 column (4 µm, 8 × 100 cm) at a flow rate of 3.0 mL/min. Preparative HPLCs were performed on a Waters LC-4000 system using a 4 × 32.5 cm C18 Delta-Pak preparative HPLC column. For normal-phase chromatographic separations, Merck grade 9385 silica gel was used with a mesh of 230-400. Rf values were determined with silica gel TLC plates (Kieselgel 60 F254, 0.25 mm layer thickness; Merck). 1H-NMR spectra were obtained on a Varian Gemini 200 spectrometer at 200 MHz using TMS as an internal standard. Fast atom bombardment mass spectra (FABMS) were obtained by M-Scan using a VG Analytical ZAB
1366
Journal of Medicinal Chemistry, 1996, Vol. 39, No. 7
Pearson et al.
2-SE instrument or by Scripps Research Institute using a VG ZAB VSE instrument. Electrospray mass spectra were obtained at Scripps Research Institute using an API III Pe Sciex triple-quadrupole mass spectrometer. S-Tritylhomocysteine. Methionine (100 g, 0.67 mol) was placed in a 5-L three-necked round bottom flask equipped with a mechanical stirrer and a cold finger condenser cooled to -78 °C with dry ice/acetone. The reaction flask was flushed with argon and cooled with a dry ice/acetone bath. Anhydrous ammonia gas was condensed in the flask until the methionine was completely dissolved. The dry ice bath was removed, and 55.7 g (2.42 mol) of sodium metal was added portionwise to the refluxing ammonia solution until the reaction mixture maintained its blue color; 47.3 g (0.61 mol) of ammonium acetate was added in portions to quench the reaction and the ammonia evaporated overnight under a stream of argon. The resulting thick solid was broken up with a spatula and then ground with a mortar and pestle. It was then combined with 182.75 g (0.70 mol) of triphenylmethanol and this mixture placed under an atmosphere of argon. With cooling by using a water bath, 500 mL of chloroform was added followed by the addition of 1000g of TFA. The reaction mixture was stirred for 2.5 h, and the solvents were removed in vacuo, the final traces being removed under high vacuum at 42 °C; 1 L of distilled water was added and the solution cooled in a water bath at 10 °C; 211.5 g of sodium hydroxide was added portionwise until a pH of 13 was achieved. The resulting white precipitate was filtered off and washed with distilled water. The solid was suspended in 1 L of water, and the pH was adjusted to 4.0 by the addition of solid citric acid; 1 L of ethyl ether was added, and the solid was filtered and washed with water followed by ether. Thoroughly drying the solid under high vacuum at 60 °C yielded 188.24 g of S-tritylhomocysteine as the product in 74.5% yield: 1H NMR (CD3OD) δ 1.88 (2H, m), 2.39 (2H, t, J ) 7.6 Hz), 3.40 (2H, t, J ) 5.6 Hz), 7.1-7.5 (15H, m); 13C NMR (CD3OD) δ 28.97, 31.62, 55.38, 67.81, 127.59, 128.71, 130.48, 145.86, 173.10. N-r-Fmoc-S-tritylhomocysteine. A three-necked round bottom flask equipped with a mechanical stirrer was charged with 108.8 g (1.03 mol) of sodium carbonate in 750 mL of water, and 161.3 g (0.425 mol) of S-tritylhomocysteine in 1 L of dioxane was added. After the solution became homogeneous, 143.7 g (0.425 mol) of N-[(9-fluorenylmethoxycarbonyl)oxy]succinimide (FmocOSu) in 1.25 L of anhydrous dioxane was added to the reaction mixture over 5 min. The reaction mixture was stirred for 3 h followed by adding citric acid until a pH of 4 was reached; 1.5 L of ethyl acetate was added, and the layers were separated in a separatory funnel. The organics were washed with saturated sodium chloride (2 × 400 mL). The solution was concentrated in vacuo to yield the crude product as an oil. This was taken up in 2 L of ethyl ether and washed with water (3 × 500 mL) and saturated NaCl (2 × 250 mL) to remove residual N-hydroxysuccinimide. The organics were dried over sodium sulfate, filtered, and concentrated in vacuo to yield an oil. The product was obtained as a foam (176.22 g, 68.7%) after further drying under high vacuum: 1H NMR (CDCl3) δ 1.53 (1H, m), 1.80 (1H, m), 2.28 (2H, m), 4.19 (3H, m), 4.39 (2H, m), 4.88 (1H, d, J ) 7.6 Hz), 7.10-7.75 (23H, m), 8.22 (1H, br s); 13C NMR (CDCl3) δ 27.90, 31.46, 47.18, 53.19, 67.18, 67.20, 119.96, 124.96, 126.72, 127.08, 127.73, 127.90, 129.60, 141.32, 143.69, 144.62, 155.99, 176.09. Preparation of N-Hydroxysuccinimide Esters. NHydroxysuccinimide esters of N-R-Cbz-N--Boc-lysine, N-RFmoc-D-tryptophan, N-Fmoc-O-tert-butyltyrosine, and N-FmocN-methylphenylalanine were prepared by reacting 1.0 equiv of amino acid with 1.1 equiv of N-hydroxysuccinimide in the presence of 1.1 equiv of diisopropylcarbodiimide in dry THF. In a representative experiment, 100 mmol of N-R-Cbz-N--Boclysine was dissolved in 250 mL of dry tetrahydrofuran (THF); 110 mmol of N-hydroxysuccinimide was added, and the reaction mixture was cooled with an ice water bath; 110 mmol of diisopropylcarbodiimide (DIC) was added, and the reaction mixture was stirred for 2 h. The progress of the reaction was monitored by thin layer chromatography (TLC). When the reaction was judged complete, the reaction mixture was
filtered through a medium frit sintered glass funnel. The collected precipitate was washed with 50 mL of dry THF, and the combined filtrates were concentrated in vacuo to yield crude product as a white paste. This paste was taken up in 200 mL of 10% ethyl acetate/ ethyl ether, and 300 mL of hexanes was added. The solution was cooled at 4 °C for 2 h. The resulting precipitate was filtered and washed twice with 100 mL of hexanes. Drying the resulting solid in vacuo yielded pure hydroxysuccinimide ester as the product in 99% yield. The product was one spot by TLC analysis in two solvent systems (Rf ) 0.95 in 4:1:1 butanol/acetic acid/water and 0.78 in 90:8:2 chloroform/ methanol/acetic acid). TLC characteristics for the other hydroxysuccinimide esters are as follows: Fmoc-D-Trp-OSu (Rf ) 0.68 in 90:8:2 chloroform/ methanol/acetic acid and 0.25 in 1:1 ethyl acetate/hexanes), Fmoc-Tyr(tBu)-OSu (Rf ) 0.76 in 90:8:2 chloroform/methanol/ acetic acid and 0.27 in 1:1 ethyl acetate/hexanes), and Fmoc(N-Me)Phe-OSu (Rf ) 0.85 in 90:8:2 chloroform/methanol/ acetic acid and 0.54 in 1:1 ethyl acetate/hexanes). Cbz-Lys(Boc)-Val-OMe (6). Valine, methyl ester, hydrochloride (17.43 g, 104.5 mmol) was reacted with diisopropylethylamine (DIEA) (33.1 mL, 190 mmol) in 500 mL dry THF. N-R-Cbz-N--Boc-lysine, hydroxysuccinimide ester (53.73 g, 95 mmol) was added portionwise, and the reaction mixture was stirred at room temperature. The progress of the reaction was monitored by TLC (product Rf ) 0.43 in 1:1 EtOAc/hexane). When the reaction mixture was judged complete, the solvents were removed in vacuo. The crude product was taken up in ethyl acetate and washed with 5% citric acid, saturated sodium bicarbonate, and saturated sodium chloride. The organic layer was dried over magnesium sulfate and filtered and the filtrate concentrated in vacuo. The resulting foam was dissolved in 300 mL of ethyl ether, and 400 mL of hexanes was added. After cooling the solution for 3 h at 4 °C, the resulting solid was filtered, washed twice with hexanes, and dried under high vacuum to yield Cbz-Lys(Boc)-Val-OMe (44.46 g, 90.07 mmol) in 95% crude yield. Fmoc-D-Trp-Lys(Boc)-Val-OMe (7). Cbz-Lys(Boc)-ValOMe (41.96 g, 85.0 mmol) was taken up in 210 mL of 1% acetic acid/methanol in a 1-L round bottom flask. The solution was purged with argon, and 500 mg of 10% palladium on carbon was added. The atmosphere of the flask was evacuated and charged with hydrogen gas. This evacuation/charging procedure was repeated three times. The reaction mixture was then stirred at room temperature under an atmosphere of hydrogen. The progress of the reaction was monitored by TLC (product Rf ) 0.17 in CHCl3/MeOH/HOAc, 90:8:2). When the starting material was no longer visible by TLC, the atmosphere of the reaction mixture was evacuated and replaced with argon. The solution was filtered through a layer of Celite, and the volatiles were removed in vacuo. The crude product was taken up in carbon tetrachloride (150 mL), and this was also removed in vacuo to remove any traces of acetic acid. Treatment with carbon tetrachloride was repeated twice more, and the crude intermediate H-Lys(Boc)-Val-OMe was dried under high vacuum for 4 h. The resulting white foam was dissolved in 200 mL of anhydrous THF, and the atmosphere of the mixture was flushed with argon gas. Fmoc-D-tryptophan, hydroxysuccinimide ester (32.0 g, 61.1 mmol) was added followed by the addition of diisopropylethylamine (29.6 mL, 170 mmol). The reaction mixture was stirred overnight under an atmosphere of argon and monitored by TLC (product Rf ) 0.51 in CHCl3/ MeOH/HOAc, 90:8:2, and 0.17 in EtOAc/hexane, 1:1). After this time the volatiles were removed in vacuo on the rotory evaporator to yield an oily residue which was taken up in 500 mL of ethyl acetate. The organics were washed with 5% citric acid, saturated sodium bicarbonate, and saturated sodium chloride followed by drying over magnesium sulfate. Filtration and concentration of the filtrate in vacuo yielded a foam which was dissolved in 300 mL of ethyl ether with 20 mL of ethyl acetate added to totally solubilize the crude product. Hexanes (400 mL) were added to the solution of crude product, and the resulting suspension was allowed to stand at 4 °C for 4 h. The solid was filtered off and washed with
99mTc-Labeled
Somatostatin Receptor-Binding Peptides
hexanes to yield the crude product as a white solid (42.9 g, 55.8 mmol) in 84.5% yield. Fmoc-Tyr(tBu)-D-Trp-Lys(Boc)-Val-OMe (8). Fmoc-DTrp-Lys(Boc)-Val-OMe (42.24 g, 55.0 mmol) was dissolved in 55 mL of dry THF in a 1-L round bottom flask equipped with a stir bar. The atmosphere of the reaction mixture was flushed with argon, and diethylamine (50 mL) was added. The reaction mixture was stirred under an atmosphere of argon at room temperature for 1 h. The volatiles were removed in vacuo on the rotory evaporator, and the intermediate H-D-TrpLys(Boc)-Val-OMe was taken up in 200 mL of acetonitrile. The volatiles were again removed in vacuo on the rotory evaporator. This acetonitrile treatment was repeated twice more to yield an oil which was taken up in 200 mL of ethyl ether and 20 mL of ethyl acetate. Hexanes (400 mL) were added, and the resulting suspension was allowed to stand at 4 °C for 4 h. The crude intermediate was filtered off and washed with hexanes to yield a solid which was dried under high vacuum for 2 h. This solid was placed in a 1-L round bottom flask equipped with a magnetic stir bar and dissolved in 150 mL of anhydrous THF, and the atmosphere of the mixture was flushed with argon gas. N-R-Fmoc-O-tert-butyltyrosine, hydroxysuccinimide ester (30.6 g, 55.0 mmol) was added followed by the addition of diisopropylethylamine (19.2 mL, 110 mmol). The reaction mixture was stirred overnight under an atmosphere of argon and monitored by TLC (product Rf ) 0.53 in CHCl3/ MeOH/HOAc, 90:8:2, and 0.14 in EtOAc/hexane, 1:1). At this time the reaction was judged to be complete, so the volatiles were removed in vacuo and the crude product was taken up in 500 mL of ethyl acetate. The organics were washed with 5% citric acid, saturated sodium bicarbonate, and saturated sodium chloride followed by drying over magnesium sulfate. Filtration and concentration of the filtrate in vacuo yielded an oil which was dissolved in 300 mL of ethyl ether with 30 mL of ethyl acetate added to totally solubilize the crude product. Hexanes (400 mL) were added to the solution of crude product, and the resulting suspension was allowed to stand at 4 °C for 4 h. The crude product was filtered off and washed with hexanes to yield a white solid (45.02 g). This was taken up in 200 mL of chloroform and applied to a column of silica gel (1000 g of slurry packed with chloroform). The column was eluted with chloroform (2.5 L) followed by 2% methanol/chloroform (7 L). Fractions (250 mL) were collected with the product eluting in fractions 5-28. Fractions containing product were combined and the volatiles removed in vacuo to yield a total of 35.74 g (36.2 mmol, 65.8% yield) of compound 8. The product was analyzed by reversed-phase HPLC (Nova-Pak column, 0-100% B/A over 10 min, 100% B for 10-15 min; solvent A ) 0.1% TFA in water, solvent B ) 0.1% TFA in 90% acetonitrile/10% water) and found to be 98% pure with a tR of 10.24 min. FABMS indicated a molecular ion peak (MH+) at 987 which corresponds to the molecular formula of C56H70N6O10 with a monoisotopic mass of 986.52. Fmoc-Phe(N-Me)-Tyr(tBu)-D-Trp-Lys(Boc)-Val-OMe (9). Fmoc-Tyr(tBu)-D-Trp-Lys(Boc)-Val-OMe (35.24 g, 35.7 mmol) was dissolved in 50 mL of dry THF in a 1-L round bottom flask equipped with a stir bar. The atmosphere of the reaction mixture was flushed with argon, and diethylamine (50 mL) was added. The reaction mixture was stirred under an atmosphere of argon at room temperature for 1 h. The volatiles were removed in vacuo on the rotory evaporator, and the intermediate H-Tyr(tBu)-D-Trp-Lys(Boc)-Val-OMe was taken up in 200 mL of acetonitrile. The solvent was again removed in vacuo on the rotory evaporator. The acetonitrile treatment was repeated twice more to yield an oil which was taken up in 400 mL of ethyl ether and 50 mL of ethyl acetate. Hexanes (400 mL) were added, and the resulting suspension was allowed to stand at 4 °C for 4 h. The crude intermediate was filtered off and washed with hexanes to yield a solid which was dried under high vacuum for 2 h. This solid (24.0 g) was placed in a 1-L round bottom flask equipped with a magnetic stir bar and dissolved in 120 mL of anhydrous THF. The atmosphere of the reaction mixture was flushed with argon gas. N-R-Fmoc-N-R-methylphenylalanine, hydroxysuccinimide ester (15.64 g, 31.4 mmol) was added
Journal of Medicinal Chemistry, 1996, Vol. 39, No. 7 1367 followed by the addition of diisopropylethylamine (10.9 mL, 62.8 mmol). The reaction mixture was stirred overnight under an atmosphere of argon and monitored by TLC (product Rf ) 0.63 in CHCl3/MeOH/HOAc, 90:8:2, and 0.20 in EtOAc/hexane, 1:1). At this time the reaction was judged to be complete, so the volatiles were removed in vacuo and the crude product was taken up in 500 mL of ethyl acetate. The organics were washed with 5% citric acid, saturated sodium bicarbonate, and saturated sodium chloride followed by drying over magnesium sulfate. Filtration and concentration of the filtrate in vacuo yielded an oil which was dissolved in 200 mL of ethyl ether with 30 mL of ethyl acetate added to totally solubilize the crude product. Hexanes (400 mL) were added to the solution of crude product, and the resulting suspension was allowed to stand at 4 °C for 4 h. The solid was filtered off and washed with hexanes to yield a white solid which was dried under high vacuum to yield crude product (34.06 g, 29.66 mmol) in 94.5% yield. Fmoc-hCys(Trt)-Phe(N-Me)-Tyr(tBu)-D-Trp-Lys(Boc)Val-OMe (10). Fmoc-Phe(N-Me)-Tyr(tBu)-D-Trp-Lys(Boc)Val-OMe (33.30 g, 29.0 mmol) was dissolved in 50 mL of dry THF in a 1-L round bottom flask equipped with a stir bar. The atmosphere of the reaction mixture was flushed with argon, and diethylamine (50 mL) was added. The reaction mixture was stirred under an atmosphere of argon at room temperature for 1 h. The volatiles were removed in vacuo on the rotory evaporator, and the intermediate H-Phe(N-Me)-Tyr(tBu)-D-Trp-Lys(Boc)-Val-OMe was taken up in 200 mL of acetonitrile. The solvent was again removed in vacuo on the rotory evaporator. The acetonitrile treatment was repeated twice more to yield an oil which was taken up in 400 mL of ethyl ether and 50 mL of ethyl acetate. Hexanes (400 mL) were added, and the resulting suspension was allowed to stand at 4 °C for 4 h. The crude intermediate was filtered off and washed with hexanes to yield a solid which was dried under high vacuum for 2 h. In a separate 1-L round bottom flask equipped with a magnetic stir bar was placed Fmoc-hCys(Trt)OH (20.30 g, 33.6 mmol) which was dissolved in 70 mL of anhydrous THF. The atmosphere of the solution was flushed with argon gas, and the solution was cooled to -15 °C with an ice/acetone bath under an atmosphere of argon. Bis(2-oxo-3-oxazolidinyl)phosphinic chloride (BOP-Cl; 10.3 g, 40.3 mmol) was added followed by the addition of diisopropylethylamine (7.02 mL, 40.3 mmol). The reaction mixture was stirred for 0.5 h at -15 °C under an atmosphere of argon. At this time the intermediate H-Phe(N-Me)-Tyr(tBu)-D-Trp-Lys(Boc)-Val-OMe formed above was added followed by the addition of 20 mL of anhydrous THF and 7.02 mL (40.3 mmol) of diisopropylethylamine. The ice bath was removed, and the reaction mixture was stirred at room temperature until the reaction was complete by TLC analysis (product Rf ) 0.57 in CHCl3/MeOH/ HOAc, 90:8:2, and 0.21 in EtOAc/hexane, 1:1). At this time the volatiles were removed in vacuo and the crude product was taken up in 300 mL of ethyl acetate. The organics were washed with 5% citric acid, saturated sodium bicarbonate, and saturated sodium chloride followed by drying over magnesium sulfate. Filtration and concentration of the filtrate in vacuo yielded an oil which was dissolved in 400 mL of ethyl ether with 40 mL of ethyl acetate added to totally solubilize the crude product. Hexanes (400 mL) were added to the solution of crude product, and the resulting suspension was allowed to stand at 4 °C for 4 h. The solid was filtered off and washed with hexanes to yield a white solid which was dried under high vacuum to give crude product (42.8 g, 28.4 mmol) in 97.8% yield. H-hCys(Trt)-Phe(N-Me)-Tyr(tBu)- D -Trp-Lys(Boc)Val-OH (11). Fmoc-hCys(Trt)-Phe(N-Me)-Tyr(tBu)-D-TrpLys(Boc)-Val-OMe (43.7 g, 29.0 mmol) was dissolved in 60 mL of dry THF in a 1-L round bottom flask equipped with a stir bar. The atmosphere of the reaction mixture was flushed with argon, and diethylamine (50 mL) was added. The reaction mixture was stirred under an atmosphere of argon at room temperature for 1 h. The volatiles were removed in vacuo on the rotory evaporator, and the intermediate H-Phe(N-Me)-Tyr-
1368
Journal of Medicinal Chemistry, 1996, Vol. 39, No. 7
Pearson et al.
(tBu)-D-Trp-Lys(Boc)-Val-OMe was taken up in 400 mL of acetonitrile. The solvent was again removed in vacuo on the rotory evaporator. The acetonitrile treatment was repeated twice more to yield an oil which was taken up in 400 mL of ethyl ether and 30 mL of ethyl acetate. Hexanes (400 mL) were added, and the resulting suspension was allowed to stand at 4 °C for 4 h. The crude intermediate was filtered off and washed with hexanes to yield a yellow solid (37.02 g, 28.8 mmol) which was dried under high vacuum for 2 h. The resulting intermediate H-hCys(Trt)-Phe(N-Me)-Trp(tBu)-D-Trp-Lys(Boc)-Val-OMe (36.64 g, 28.5 mmol) was dissolved in 50 mL of dry THF in a 1-L round bottom flask. The atmosphere was flushed with argon gas, and a solution of 1.44 g of lithium hydroxide hydrate (34.2 mmol) dissolved in 10 mL of water was added. This was followed by the addition of another 80 mL of THF to achieve solution homogeneity. The reaction mixture was stirred under argon for 4 h at room temperature. Reversed-phase HPLC analysis (Nova-Pak column, 0-100% B/A over 10 min, 100% B for 10-15 min; solvent A ) 0.1% TFA in water, solvent B ) 0.1% TFA in 90% acetonitrile/10% water) showed that the reaction was incomplete (methyl ester tR ) 10.16 min, free carboxyl tR ) 10.06), so an additional 0.94 g of lithium hydroxide (22.3 mmol) in 10 mL of water was added followed by the addition of another 50 mL of THF. The reaction mixture was stirred overnight at which time HPLC analysis indicated that the reaction was complete. At this time the volatiles were removed in vacuo and the crude product was taken up in 400 mL of ethyl acetate. The organics were washed with 5% citric acid and saturated sodium chloride followed by drying over magnesium sulfate. Filtration and concentration of the filtrate in vacuo yielded an oil which was dissolved in 300 mL of ethyl ether with 50 mL of ethyl acetate added to totally solubilize the crude product. Hexanes (400 mL) were added to the solution of crude product, and the resulting suspension was allowed to stand at 4 °C for 4 h. The solid was filtered off and washed with hexanes to yield a white solid which was dried under high vacuum to yield 35.7 g (28.05 mmol) of crude product in 98.4% yield. cyclo-[hCys(Trt)-Phe(N-Me)-Tyr(tBu)-D-Trp-Lys(Boc)Val] (12). H-hCys(Trt)-Phe(N-Me)-Trp(tBu)-D-Trp-Lys(Boc)Val-OH (35.0 g, 27.52 mmol) was dissolved in 3500 mL of anhydrous dimethylformamide (DMF) in a dry 5-L round bottom flask. The atmosphere was flushed with argon gas, and 61.2 mL (27.54 mmol) of a 0.45 M solution of 1:1 2-(1Hbenzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate (HBTU)/N-hydroxybenzotriazole (HOBt) was added via syringe. The reaction mixture was stirred for 0.5 h at room temperature under an atmosphere of argon, and 9.60 mL (55.04 mmol) of diisopropylethylamine was added via syringe. The reaction mixture was stirred at room temperature until reversed-phase HPLC analysis (Nova-Pak column, 0-100% B/A over 10 min, 100% B for 10-15 min; solvent A ) 0.1% TFA in water, solvent B ) 0.1% TFA in 90% acetonitrile/10% water) indicated that the starting material (tR ) 10.06 min) had disappeared to be replaced by product (tR ) 10.40 min). At this time the DMF was removed in vacuo to yield crude product as an oil. This was taken up in 400 mL of ether and 50 mL of ethyl acetate. Hexanes (400 mL) were added, and the solution was left standing for 4 h. The solution was decanted from the resulting gum which was triturated with hexanes until the residue solidified. The solid was collected by filtration and washed with hexanes to yield crude product (33.8 g, 26.96 mmol) in 98% yield after drying under high vacuum overnight. cyclo-[Homocysteinyl-N-methylphenylalanyl-tyrosylD-tryptophanyl-lysyl-valinyl] (13). A deprotection cocktail consisting of 200 mL of trifluoroacetic acid, 10 mL of water, 5 mL of ethanedithiol, and 4 mL of triisopropylsilane was prepared and added to cyclo-[hCys(Trt)-Phe(N-Me)-Trp(tBu)D-Trp-Lys(Boc)-Val] (33.5 g, 26.7 mmol) in a 1-L round bottom flask. The reaction mixture was stirred for 1 h at room temperature at which time the mixture turned from red to a light yellow. The deprotection mixture was added slowly with stirring to 1200 mL of cold ethyl ether in a 2-L Ehrlenmeyer
flask. After addition was complete, the flask was sealed and left standing at 4 °C for 4 h to complete the precipitation of crude product. The solid was collected by filtration and washed with cold ether to yield 29.72 g of crude product after drying under high vacuum overnight. The product was purified in 15 2.0-g portions. Therefore, 2.0 g of crude peptide 10 was dissolved in 10 mL of 0.1% TFA in 1:1 acetonitrile/water, and this solution was filtered through a 0.2-µm nylon filter. The filter was washed with an additional 2 mL of 0.1% TFA in 1:1 acetonitrile/water and the total filtrate (12 mL volume) loaded directly onto a 4 × 32.5 cm C18 Delta-Pak preparative HPLC column running in 0.1% TFA/water (solvent A) at 25 mL/min. After the crude product had been completely loaded, the flow rate was increased to 75 mL/min. After 5 min, the ratio of solvent B (0.1% TFA in 90% acetonitrile/10% water) was increased from 0% to 10%. After 5 min at 10% B, solvent B was increased from 10% to 20%. After 5 min at 20% B, solvent B was increased from 20 to 30%. After 5 min at 30% B, solvent B was increased linearly from 30% to 50% over 20 min. Solvent B was then increased to 100% and held at 100% for an additional 10 min. Fractions were collected based on effluent monitoring at 230 nm. Collected fractions were evaluated by analytical HPLC (Delta-Pak column, 30-50% B/A over 20 min, 100% B for 2025 min; solvent A ) 0.1% TFA in water, solvent B ) 0.1% TFA in 90% acetonitrile/10% water). Fractions containing 13 (tR ) 11.28 min) with a purity of g95% were pooled, the acetonitrile was evaporated in vacuo, and the final aqueous solution was lyophilized to yield solid 13 as the trifluoroacetate salt. Fractions containing peptide 13 with a purity of
